LLM Observability & Explainability

Real-time monitoring of LLM usage and AI-judged quality metrics.

Target LLM (Responder)

Total Number of Calls
Not Applicable
Average Token Usage
Not Applicable
Average Response Rate
Not Applicable
Average Latency
Not Applicable
Average Accuracy
Not Applicable
Average Relevance
Not Applicable
Average Reasoning Quality
Not Applicable
Average Hallucination
Not Applicable

History

Group data by

Prompt evaluation

Co-funded by the European Union (COP-PILOT, 101189819) and the Swiss State Secretariat for Education, Research and Innovation (SERI)